internal.h 16.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10
/* internal.h: mm/ internal definitions
 *
 * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
 * Written by David Howells (dhowells@redhat.com)
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; either version
 * 2 of the License, or (at your option) any later version.
 */
11 12 13
#ifndef __MM_INTERNAL_H
#define __MM_INTERNAL_H

14
#include <linux/fs.h>
15
#include <linux/mm.h>
16
#include <linux/pagemap.h>
17
#include <linux/tracepoint-defs.h>
L
Linus Torvalds 已提交
18

19 20 21 22 23 24 25
/*
 * The set of flags that only affect watermark checking and reclaim
 * behaviour. This is used by the MM to obey the caller constraints
 * about IO, FS and watermark checking while ignoring placement
 * hints such as HIGHMEM usage.
 */
#define GFP_RECLAIM_MASK (__GFP_RECLAIM|__GFP_HIGH|__GFP_IO|__GFP_FS|\
26
			__GFP_NOWARN|__GFP_RETRY_MAYFAIL|__GFP_NOFAIL|\
27 28
			__GFP_NORETRY|__GFP_MEMALLOC|__GFP_NOMEMALLOC|\
			__GFP_ATOMIC)
29 30 31 32 33 34 35 36 37 38

/* The GFP flags allowed during early boot */
#define GFP_BOOT_MASK (__GFP_BITS_MASK & ~(__GFP_RECLAIM|__GFP_IO|__GFP_FS))

/* Control allocation cpuset and node placement constraints */
#define GFP_CONSTRAINT_MASK (__GFP_HARDWALL|__GFP_THISNODE)

/* Do not use these with a slab allocator */
#define GFP_SLAB_BUG_MASK (__GFP_DMA32|__GFP_HIGHMEM|~__GFP_BITS_MASK)

39 40
void page_writeback_init(void);

J
Jan Kara 已提交
41
int do_swap_page(struct vm_fault *vmf);
42

43 44 45
void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *start_vma,
		unsigned long floor, unsigned long ceiling);

46 47 48 49 50
static inline bool can_madv_dontneed_vma(struct vm_area_struct *vma)
{
	return !(vma->vm_flags & (VM_LOCKED|VM_HUGETLB|VM_PFNMAP));
}

M
Michal Hocko 已提交
51 52 53 54 55
void unmap_page_range(struct mmu_gather *tlb,
			     struct vm_area_struct *vma,
			     unsigned long addr, unsigned long end,
			     struct zap_details *details);

56 57 58 59 60 61 62 63 64 65 66 67 68 69
extern int __do_page_cache_readahead(struct address_space *mapping,
		struct file *filp, pgoff_t offset, unsigned long nr_to_read,
		unsigned long lookahead_size);

/*
 * Submit IO for the read-ahead request in file_ra_state.
 */
static inline unsigned long ra_submit(struct file_ra_state *ra,
		struct address_space *mapping, struct file *filp)
{
	return __do_page_cache_readahead(mapping, filp,
					ra->start, ra->size, ra->async_size);
}

70
/*
71
 * Turn a non-refcounted page (->_refcount == 0) into refcounted with
72 73 74 75
 * a count of one.
 */
static inline void set_page_refcounted(struct page *page)
{
76
	VM_BUG_ON_PAGE(PageTail(page), page);
77
	VM_BUG_ON_PAGE(page_ref_count(page), page);
N
Nick Piggin 已提交
78 79 80
	set_page_count(page, 1);
}

H
Hugh Dickins 已提交
81 82
extern unsigned long highest_memmap_pfn;

83 84 85 86 87 88
/*
 * Maximum number of reclaim retries without progress before the OOM
 * killer is consider the only way forward.
 */
#define MAX_RECLAIM_RETRIES 16

L
Lee Schermerhorn 已提交
89 90 91
/*
 * in mm/vmscan.c:
 */
92
extern int isolate_lru_page(struct page *page);
L
Lee Schermerhorn 已提交
93
extern void putback_lru_page(struct page *page);
94

B
Bob Liu 已提交
95 96 97 98 99
/*
 * in mm/rmap.c:
 */
extern pmd_t *mm_find_pmd(struct mm_struct *mm, unsigned long address);

L
Lee Schermerhorn 已提交
100 101 102
/*
 * in mm/page_alloc.c
 */
103

104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
/*
 * Structure for holding the mostly immutable allocation parameters passed
 * between functions involved in allocations, including the alloc_pages*
 * family of functions.
 *
 * nodemask, migratetype and high_zoneidx are initialized only once in
 * __alloc_pages_nodemask() and then never change.
 *
 * zonelist, preferred_zone and classzone_idx are set first in
 * __alloc_pages_nodemask() for the fast path, and might be later changed
 * in __alloc_pages_slowpath(). All other functions pass the whole strucure
 * by a const pointer.
 */
struct alloc_context {
	struct zonelist *zonelist;
	nodemask_t *nodemask;
120
	struct zoneref *preferred_zoneref;
121 122
	int migratetype;
	enum zone_type high_zoneidx;
123
	bool spread_dirty_pages;
124 125
};

126 127
#define ac_classzone_idx(ac) zonelist_zone_idx(ac->preferred_zoneref)

128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145
/*
 * Locate the struct page for both the matching buddy in our
 * pair (buddy1) and the combined O(n+1) page they form (page).
 *
 * 1) Any buddy B1 will have an order O twin B2 which satisfies
 * the following equation:
 *     B2 = B1 ^ (1 << O)
 * For example, if the starting buddy (buddy2) is #8 its order
 * 1 buddy is #10:
 *     B2 = 8 ^ (1 << 1) = 8 ^ 2 = 10
 *
 * 2) Any buddy B will have an order O+1 parent P which
 * satisfies the following equation:
 *     P = B & ~(1 << O)
 *
 * Assumption: *_mem_map is contiguous at least up to MAX_ORDER
 */
static inline unsigned long
146
__find_buddy_pfn(unsigned long page_pfn, unsigned int order)
147
{
148
	return page_pfn ^ (1 << order);
149 150
}

151 152 153 154 155 156 157 158 159 160 161 162
extern struct page *__pageblock_pfn_to_page(unsigned long start_pfn,
				unsigned long end_pfn, struct zone *zone);

static inline struct page *pageblock_pfn_to_page(unsigned long start_pfn,
				unsigned long end_pfn, struct zone *zone)
{
	if (zone->contiguous)
		return pfn_to_page(start_pfn);

	return __pageblock_pfn_to_page(start_pfn, end_pfn, zone);
}

163
extern int __isolate_free_page(struct page *page, unsigned int order);
164 165
extern void __free_pages_bootmem(struct page *page, unsigned long pfn,
					unsigned int order);
166
extern void prep_compound_page(struct page *page, unsigned int order);
167 168
extern void post_alloc_hook(struct page *page, unsigned int order,
					gfp_t gfp_flags);
169
extern int user_min_free_kbytes;
170

171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
#if defined CONFIG_COMPACTION || defined CONFIG_CMA

/*
 * in mm/compaction.c
 */
/*
 * compact_control is used to track pages being migrated and the free pages
 * they are being migrated to during memory compaction. The free_pfn starts
 * at the end of a zone and migrate_pfn begins at the start. Movable pages
 * are moved to the end of a zone during a compaction run and the run
 * completes when free_pfn <= migrate_pfn
 */
struct compact_control {
	struct list_head freepages;	/* List of free pages to migrate to */
	struct list_head migratepages;	/* List of pages being migrated */
186
	struct zone *zone;
187 188
	unsigned long nr_freepages;	/* Number of isolated free pages */
	unsigned long nr_migratepages;	/* Number of pages to migrate */
189 190
	unsigned long total_migrate_scanned;
	unsigned long total_free_scanned;
191 192
	unsigned long free_pfn;		/* isolate_freepages search base */
	unsigned long migrate_pfn;	/* isolate_migratepages search base */
193
	unsigned long last_migrated_pfn;/* Not yet flushed page being freed */
194 195
	const gfp_t gfp_mask;		/* gfp mask of a direct compactor */
	int order;			/* order a direct compactor needs */
196
	int migratetype;		/* migratetype of direct compactor */
197 198
	const unsigned int alloc_flags;	/* alloc flags of a direct compactor */
	const int classzone_idx;	/* zone index of a direct compactor */
199
	enum migrate_mode mode;		/* Async or sync migration mode */
200
	bool ignore_skip_hint;		/* Scan blocks even if marked skip */
201
	bool no_set_skip_hint;		/* Don't mark blocks for skipping */
202
	bool ignore_block_suitable;	/* Scan blocks considered unsuitable */
203
	bool direct_compaction;		/* False from kcompactd or /proc/... */
204
	bool whole_zone;		/* Whole zone should/has been scanned */
205
	bool contended;			/* Signal lock or sched contention */
206
	bool finishing_block;		/* Finishing current pageblock */
207 208 209
};

unsigned long
210 211
isolate_freepages_range(struct compact_control *cc,
			unsigned long start_pfn, unsigned long end_pfn);
212
unsigned long
213 214
isolate_migratepages_range(struct compact_control *cc,
			   unsigned long low_pfn, unsigned long end_pfn);
215 216
int find_suitable_fallback(struct free_area *area, unsigned int order,
			int migratetype, bool only_stealable, bool *can_steal);
217 218

#endif
219

220
/*
221 222 223 224
 * This function returns the order of a free page in the buddy system. In
 * general, page_zone(page)->lock must be held by the caller to prevent the
 * page from being allocated in parallel and returning garbage as the order.
 * If a caller does not hold page_zone(page)->lock, it must guarantee that the
225 226
 * page cannot be allocated or merged in parallel. Alternatively, it must
 * handle invalid values gracefully, and use page_order_unsafe() below.
227
 */
228
static inline unsigned int page_order(struct page *page)
229
{
230
	/* PageBuddy() must be checked by the caller */
231 232
	return page_private(page);
}
233

234 235 236 237 238
/*
 * Like page_order(), but for callers who cannot afford to hold the zone lock.
 * PageBuddy() should be checked first by the caller to minimize race window,
 * and invalid values must be handled gracefully.
 *
239
 * READ_ONCE is used so that if the caller assigns the result into a local
240 241 242 243 244
 * variable and e.g. tests it for valid range before using, the compiler cannot
 * decide to remove the variable and inline the page_private(page) multiple
 * times, potentially observing different values in the tests and the actual
 * use of the result.
 */
245
#define page_order_unsafe(page)		READ_ONCE(page_private(page))
246

247 248 249 250 251
static inline bool is_cow_mapping(vm_flags_t flags)
{
	return (flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE;
}

252 253 254 255 256 257 258
/*
 * These three helpers classifies VMAs for virtual memory accounting.
 */

/*
 * Executable code area - executable, not writable, not stack
 */
259 260
static inline bool is_exec_mapping(vm_flags_t flags)
{
261
	return (flags & (VM_EXEC | VM_WRITE | VM_STACK)) == VM_EXEC;
262 263
}

264 265 266 267 268 269
/*
 * Stack area - atomatically grows in one direction
 *
 * VM_GROWSUP / VM_GROWSDOWN VMAs are always private anonymous:
 * do_mmap() forbids all other combinations.
 */
270 271
static inline bool is_stack_mapping(vm_flags_t flags)
{
272
	return (flags & VM_STACK) == VM_STACK;
273 274
}

275 276 277
/*
 * Data area - private, writable, not stack
 */
278 279
static inline bool is_data_mapping(vm_flags_t flags)
{
280
	return (flags & (VM_WRITE | VM_SHARED | VM_STACK)) == VM_WRITE;
281 282
}

283 284 285 286
/* mm/util.c */
void __vma_link_list(struct mm_struct *mm, struct vm_area_struct *vma,
		struct vm_area_struct *prev, struct rb_node *rb_parent);

H
Hugh Dickins 已提交
287
#ifdef CONFIG_MMU
288
extern long populate_vma_page_range(struct vm_area_struct *vma,
289
		unsigned long start, unsigned long end, int *nonblocking);
H
Hugh Dickins 已提交
290 291 292 293 294 295 296
extern void munlock_vma_pages_range(struct vm_area_struct *vma,
			unsigned long start, unsigned long end);
static inline void munlock_vma_pages_all(struct vm_area_struct *vma)
{
	munlock_vma_pages_range(vma, vma->vm_start, vma->vm_end);
}

N
Nick Piggin 已提交
297
/*
H
Hugh Dickins 已提交
298
 * must be called with vma's mmap_sem held for read or write, and page locked.
N
Nick Piggin 已提交
299 300
 */
extern void mlock_vma_page(struct page *page);
301
extern unsigned int munlock_vma_page(struct page *page);
N
Nick Piggin 已提交
302 303 304 305 306 307 308 309 310 311

/*
 * Clear the page's PageMlocked().  This can be useful in a situation where
 * we want to unconditionally remove a page from the pagecache -- e.g.,
 * on truncation or freeing.
 *
 * It is legal to call this function for any page, mlocked or not.
 * If called for a page that is still mapped by mlocked vmas, all we do
 * is revert to lazy LRU behaviour -- semantics are not broken.
 */
312
extern void clear_page_mlock(struct page *page);
N
Nick Piggin 已提交
313 314

/*
315 316 317
 * mlock_migrate_page - called only from migrate_misplaced_transhuge_page()
 * (because that does not go through the full procedure of migration ptes):
 * to migrate the Mlocked page flag; update statistics.
N
Nick Piggin 已提交
318 319 320
 */
static inline void mlock_migrate_page(struct page *newpage, struct page *page)
{
N
Nick Piggin 已提交
321
	if (TestClearPageMlocked(page)) {
322
		int nr_pages = hpage_nr_pages(page);
N
Nick Piggin 已提交
323

324
		/* Holding pmd lock, no change in irq context: __mod is safe */
325
		__mod_zone_page_state(page_zone(page), NR_MLOCK, -nr_pages);
N
Nick Piggin 已提交
326
		SetPageMlocked(newpage);
327
		__mod_zone_page_state(page_zone(newpage), NR_MLOCK, nr_pages);
N
Nick Piggin 已提交
328
	}
N
Nick Piggin 已提交
329 330
}

331 332
extern pmd_t maybe_pmd_mkwrite(pmd_t pmd, struct vm_area_struct *vma,
		bool dirty);
333

334 335 336 337 338 339 340 341 342 343 344 345 346
/*
 * At what user virtual address is page expected in @vma?
 */
static inline unsigned long
__vma_address(struct page *page, struct vm_area_struct *vma)
{
	pgoff_t pgoff = page_to_pgoff(page);
	return vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
}

static inline unsigned long
vma_address(struct page *page, struct vm_area_struct *vma)
{
347 348 349 350
	unsigned long start, end;

	start = __vma_address(page, vma);
	end = start + PAGE_SIZE * (hpage_nr_pages(page) - 1);
351 352

	/* page should be within @vma mapping range */
353
	VM_BUG_ON_VMA(end < vma->vm_start || start >= vma->vm_end, vma);
354

355
	return max(start, vma->vm_start);
356 357
}

H
Hugh Dickins 已提交
358
#else /* !CONFIG_MMU */
N
Nick Piggin 已提交
359 360 361 362
static inline void clear_page_mlock(struct page *page) { }
static inline void mlock_vma_page(struct page *page) { }
static inline void mlock_migrate_page(struct page *new, struct page *old) { }

H
Hugh Dickins 已提交
363
#endif /* !CONFIG_MMU */
L
Lee Schermerhorn 已提交
364

365 366 367 368 369 370 371 372
/*
 * Return the mem_map entry representing the 'offset' subpage within
 * the maximally aligned gigantic page 'base'.  Handle any discontiguity
 * in the mem_map at MAX_ORDER_NR_PAGES boundaries.
 */
static inline struct page *mem_map_offset(struct page *base, int offset)
{
	if (unlikely(offset >= MAX_ORDER_NR_PAGES))
F
Fabian Frederick 已提交
373
		return nth_page(base, offset);
374 375 376 377
	return base + offset;
}

/*
L
Lucas De Marchi 已提交
378
 * Iterator over all subpages within the maximally aligned gigantic
379 380 381 382 383 384 385 386 387 388 389 390 391 392
 * page 'base'.  Handle any discontiguity in the mem_map.
 */
static inline struct page *mem_map_next(struct page *iter,
						struct page *base, int offset)
{
	if (unlikely((offset & (MAX_ORDER_NR_PAGES - 1)) == 0)) {
		unsigned long pfn = page_to_pfn(base) + offset;
		if (!pfn_valid(pfn))
			return NULL;
		return pfn_to_page(pfn);
	}
	return iter + 1;
}

393 394 395 396 397 398 399 400 401 402 403 404
/*
 * FLATMEM and DISCONTIGMEM configurations use alloc_bootmem_node,
 * so all functions starting at paging_init should be marked __init
 * in those cases. SPARSEMEM, however, allows for memory hotplug,
 * and alloc_bootmem_node is not used.
 */
#ifdef CONFIG_SPARSEMEM
#define __paginginit __meminit
#else
#define __paginginit __init
#endif

405 406 407 408 409 410 411 412 413 414 415 416 417 418
/* Memory initialisation debug and verification */
enum mminit_level {
	MMINIT_WARNING,
	MMINIT_VERIFY,
	MMINIT_TRACE
};

#ifdef CONFIG_DEBUG_MEMORY_INIT

extern int mminit_loglevel;

#define mminit_dprintk(level, prefix, fmt, arg...) \
do { \
	if (level < mminit_loglevel) { \
419
		if (level <= MMINIT_WARNING) \
420
			pr_warn("mminit::" prefix " " fmt, ##arg);	\
421 422
		else \
			printk(KERN_DEBUG "mminit::" prefix " " fmt, ##arg); \
423 424 425
	} \
} while (0)

426
extern void mminit_verify_pageflags_layout(void);
427
extern void mminit_verify_zonelist(void);
428 429 430 431 432 433 434
#else

static inline void mminit_dprintk(enum mminit_level level,
				const char *prefix, const char *fmt, ...)
{
}

435 436 437 438
static inline void mminit_verify_pageflags_layout(void)
{
}

439 440 441
static inline void mminit_verify_zonelist(void)
{
}
442
#endif /* CONFIG_DEBUG_MEMORY_INIT */
443 444 445 446 447 448 449 450 451 452 453 454

/* mminit_validate_memmodel_limits is independent of CONFIG_DEBUG_MEMORY_INIT */
#if defined(CONFIG_SPARSEMEM)
extern void mminit_validate_memmodel_limits(unsigned long *start_pfn,
				unsigned long *end_pfn);
#else
static inline void mminit_validate_memmodel_limits(unsigned long *start_pfn,
				unsigned long *end_pfn)
{
}
#endif /* CONFIG_SPARSEMEM */

455 456 457 458
#define NODE_RECLAIM_NOSCAN	-2
#define NODE_RECLAIM_FULL	-1
#define NODE_RECLAIM_SOME	0
#define NODE_RECLAIM_SUCCESS	1
W
Wu Fengguang 已提交
459

460 461
extern int hwpoison_filter(struct page *p);

W
Wu Fengguang 已提交
462 463
extern u32 hwpoison_filter_dev_major;
extern u32 hwpoison_filter_dev_minor;
W
Wu Fengguang 已提交
464 465
extern u64 hwpoison_filter_flags_mask;
extern u64 hwpoison_filter_flags_value;
A
Andi Kleen 已提交
466
extern u64 hwpoison_filter_memcg;
467
extern u32 hwpoison_filter_enable;
A
Al Viro 已提交
468

469
extern unsigned long  __must_check vm_mmap_pgoff(struct file *, unsigned long,
A
Al Viro 已提交
470
        unsigned long, unsigned long,
M
Michal Hocko 已提交
471
        unsigned long, unsigned long);
472 473

extern void set_pageblock_order(void);
474 475
unsigned long reclaim_clean_pages_from_list(struct zone *zone,
					    struct list_head *page_list);
476 477 478 479 480 481 482 483 484
/* The ALLOC_WMARK bits are used as an index to zone->watermark */
#define ALLOC_WMARK_MIN		WMARK_MIN
#define ALLOC_WMARK_LOW		WMARK_LOW
#define ALLOC_WMARK_HIGH	WMARK_HIGH
#define ALLOC_NO_WATERMARKS	0x04 /* don't check watermarks at all */

/* Mask to get the watermark bits */
#define ALLOC_WMARK_MASK	(ALLOC_NO_WATERMARKS-1)

485 486 487 488 489 490 491 492 493 494 495
/*
 * Only MMU archs have async oom victim reclaim - aka oom_reaper so we
 * cannot assume a reduced access to memory reserves is sufficient for
 * !MMU
 */
#ifdef CONFIG_MMU
#define ALLOC_OOM		0x08
#else
#define ALLOC_OOM		ALLOC_NO_WATERMARKS
#endif

496 497 498 499 500
#define ALLOC_HARDER		0x10 /* try to alloc harder */
#define ALLOC_HIGH		0x20 /* __GFP_HIGH set */
#define ALLOC_CPUSET		0x40 /* check for correct cpuset */
#define ALLOC_CMA		0x80 /* allow allocations from CMA areas */

501 502 503
enum ttu_flags;
struct tlbflush_unmap_batch;

504 505 506 507 508 509 510

/*
 * only for MM internal work items which do not depend on
 * any allocations or locks which might depend on allocations
 */
extern struct workqueue_struct *mm_percpu_wq;

511 512
#ifdef CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
void try_to_unmap_flush(void);
513
void try_to_unmap_flush_dirty(void);
514
void flush_tlb_batched_pending(struct mm_struct *mm);
515 516 517 518
#else
static inline void try_to_unmap_flush(void)
{
}
519 520 521
static inline void try_to_unmap_flush_dirty(void)
{
}
522 523 524
static inline void flush_tlb_batched_pending(struct mm_struct *mm)
{
}
525
#endif /* CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH */
526 527 528 529 530

extern const struct trace_print_flags pageflag_names[];
extern const struct trace_print_flags vmaflag_names[];
extern const struct trace_print_flags gfpflag_names[];

531 532 533 534 535 536 537 538 539 540
static inline bool is_migrate_highatomic(enum migratetype migratetype)
{
	return migratetype == MIGRATE_HIGHATOMIC;
}

static inline bool is_migrate_highatomic_page(struct page *page)
{
	return get_pageblock_migratetype(page) == MIGRATE_HIGHATOMIC;
}

541
void setup_zone_pageset(struct zone *zone);
542
#endif	/* __MM_INTERNAL_H */